Accuracy In Toric Intraocular Lens Repositioning: A Comparison Of Current Realignment Calculation Methods

Published 2026 - 44th Congress of the ESCRS

Reference: FP05.05 | Type: Free paper | DOI: 10.82333/ejp0-je52

Authors: Dr Christoph Lwowski* ¹ , Dr Klemens Kaiser ² , Ms Mariam Sobhi ² , Prof. Dr Thomas Kohnen ²

¹Ophthalmology,University Clinic Frankfurt,Frankfurt am Main ,Germany;Ophthalmology,University Clinic Lübeck,Lübeck ,Germany, ²Ophthalmology,University Clinic Frankfurt,Frankfurt am Main ,Germany

Purpose

To compare different calculation methods for toric intraocular lens (IOL) realignment in eyes with postoperative residual astigmatism.

Setting

Retrospective cohort study.

Methods

49 eyes of 49 patients undergoing toric IOL rotation at a single tertiary referral center. Predicted postoperative astigmatism and recommended rotation axes were calculated using AstiFix, Gatinel, Ophtera, Savini, and IOLCon. Prediction error (PE) was evaluated using vector magnitude and bivariate vector analysis using the Wilcox-Holladay-Wang-Koch statistical software implemented in R. Agreement between axis recommendations was assessed using intraclass correlation-coefficients (ICC). Calculations were performed using phakic (IOLMaster 700 and Pentacam) and pseudophakic (Pentacam) biometric and tomographic input data where applicable.

Results

AstiFix, Gatinel, and Ophtera showed similar centroids and polygon distributions, with low PE magnitudes (0.43D, 0.45D, and 0.46D, respectively)

and near-zero bivariate bias. No statistically significant differences in vector PE were observed among these three tools. Savini demonstrated significantly different axis recommendations compared with AstiFix, Gatinel, and Ophtera, while IOLCon showed a significant bivariate bias relative to the zero vector. Axis agreement between AstiFix, Gatinel, and Ophtera was excellent (ICC=0.999), whereas agreement with Savini was lower (ICC=0.903). The use of repeated biometric measurements of the pseudophakic (pre-rotational) eye did not result in a statistically significant improvement in PE.

Conclusions

Realignment formulas incorporating pre-rotational subjective refraction provide consistent and accurate predictions for toric IOL rotation. Tools primarily designed for toric IOL power calculationmay be less suitable for realignment planning. Subjective refraction remains the key determinant for accurate realignment planning.